NeurIPS 2016

574 papers

Top Research Directions

Research directions most represented by NeurIPS 2016 papers.

Scaling Gaussian Processes Beyond Cubic Complexity 9
The Statistical Computational Gap in High Dimensions 8
Learning Densities with Intractable Normalizing Constants 8
Kernel Hyperparameter Sensitivity in Distributional Metrics 7
Amortized Bayesian Inference with Neural Networks 6
Basins of Attraction in Nonconvex Optimization 6
Leverage Score Sampling for Scalable Matrix Computations 5
Non-linear Modeling of Neural Population Dynamics 5
Semidefinite Relaxation for Non-Convex Engineering Optimization 5
Inducing Point Approximations for Scalable Gaussian Processes 5
Variance Reduction in Stochastic Optimization 5
Overcoming the Exponential Bottleneck in Tensor Decompositions 5
Overcoming the Cubic Complexity of Kernel Methods 5
Generalizing Matrix and Tensor Recovery to Asymmetric Systems 4
Beyond the Independent Arm Assumption in Bandit Learning 4

Top Authors

AuthorPapers
Barnabas Poczos6
Inderjit S Dhillon6
Nathan Srebro6
Pieter Abbeel6
Russ Salakhutdinov6
Andreas Krause5
Francis Bach5
Koray Kavukcuoglu5
Sergey Levine5
Zhaoran Wang5
Alex Graves4
Han Liu4
John Langford4
Joshua B. Tenenbaum4
Lawrence Carin4
Oriol Vinyals4
Prateek Jain4
Remi Munos4
Stefano Ermon4
Suvrit Sra4

Impactful Papers

Papers from NeurIPS 2016 that shaped research clusters.

Foundational

Influential

Papers

Title
Stochastic Variational Deep Kernel Learning
Strategic Attentive Writer for Learning Macro-Actions
Structure-Blind Signal Recovery
Structured Matrix Recovery via the Generalized Dantzig Selector
Structured Prediction Theory Based on Factor Graph Complexity
Structured Sparse Regression via Greedy Hard Thresholding
Sublinear Time Orthogonal Tensor Decomposition
Sub-sampled Newton Methods with Non-uniform Sampling
Supervised learning through the lens of compression
Supervised Learning with Tensor Networks
Supervised Word Mover's Distance
SURGE: Surface Regularized Geometry Estimation from a Single Image
Swapout: Learning an ensemble of deep architectures
Synthesis of MCMC and Belief Propagation
Synthesizing the preferred inputs for neurons in neural networks via deep generator networks
Tagger: Deep Unsupervised Perceptual Grouping
Temporal Regularized Matrix Factorization for High-dimensional Time Series Prediction
Tensor Switching Networks
Testing for Differences in Gaussian Graphical Models: Applications to Brain Connectivity
The Forget-me-not Process
The Generalized Reparameterization Gradient
The Limits of Learning with Missing Data
The Multi-fidelity Multi-armed Bandit
The Multiple Quantile Graphical Model
The Multiscale Laplacian Graph Kernel
The non-convex Burer-Monteiro approach works on smooth semidefinite programs
Theoretical Comparisons of Positive-Unlabeled Learning against Positive-Negative Learning
The Parallel Knowledge Gradient Method for Batch Bayesian Optimization
The Power of Adaptivity in Identifying Statistical Alternatives
The Power of Optimization from Samples
The Product Cut
The Robustness of Estimator Composition
The Sound of APALM Clapping: Faster Nonsmooth Nonconvex Optimization with Stochastic Asynchronous PALM
Threshold Bandits, With and Without Censored Feedback
Threshold Learning for Optimal Decision Making
Tight Complexity Bounds for Optimizing Composite Objectives
Total Variation Classes Beyond 1d: Minimax Rates, and the Limitations of Linear Smoothers
Toward Deeper Understanding of Neural Networks: The Power of Initialization and a Dual View on Expressivity
Towards Conceptual Compression
Towards Unifying Hamiltonian Monte Carlo and Slice Sampling
Tracking the Best Expert in Non-stationary Stochastic Environments
Tractable Operations for Arithmetic Circuits of Probabilistic Models
Training and Evaluating Multimodal Word Embeddings with Large-scale Web Annotated Images
Tree-Structured Reinforcement Learning for Sequential Object Localization
Truncated Variance Reduction: A Unified Approach to Bayesian Optimization and Level-Set Estimation
Understanding Probabilistic Sparse Gaussian Process Approximations
Understanding the Effective Receptive Field in Deep Convolutional Neural Networks
Unified Methods for Exploiting Piecewise Linear Structure in Convex Optimization
Unifying Count-Based Exploration and Intrinsic Motivation
Universal Correspondence Network

« prev Page 11/12 next »